adds data analysis

app/analysis.py

@@ -0,0 +1,60 @@
+import os
+import pandas as pd
+import matplotlib
+matplotlib.use("Agg")  # Prevents GUI-related issues in Flask
+
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+
+def load_data(file_path: str) -> pd.DataFrame:
+    """Loads the scraped data from a CSV file into a Pandas DataFrame."""
+    if not os.path.exists(file_path):
+        raise FileNotFoundError(f"File {file_path} not found.")
+
+    df = pd.read_csv(file_path)
+    df["timestamp"] = pd.to_datetime(df["timestamp"], errors="coerce")
+    df["last_action"] = pd.to_datetime(df["last_action"], errors="coerce")
+
+    return df
+
+def generate_statistics(df: pd.DataFrame):
+    """Generates activity statistics grouped by hour."""
+    df["hour"] = df["timestamp"].dt.hour
+    return df.groupby("hour").size()
+
+def plot_activity_distribution(df: pd.DataFrame, output_path="activity_distribution.png"):
+    """Plots user activity distribution and saves the figure."""
+
+    # Ensure the directory exists
+    static_dir = os.path.join("app", "static", "plots")
+    output_path = os.path.join(static_dir, output_path)
+    os.makedirs(static_dir, exist_ok=True)
+
+    # Convert timestamp column to datetime (if not already)
+    if not pd.api.types.is_datetime64_any_dtype(df["timestamp"]):
+        df["timestamp"] = pd.to_datetime(df["timestamp"], errors="coerce")
+
+    df["hour"] = df["timestamp"].dt.hour
+    activity_counts = df.groupby("hour").size().reset_index(name="count")
+
+    # Use non-GUI backend for Matplotlib
+    plt.figure(figsize=(10, 5))
+
+    # Fix Seaborn Warning: Assign `hue` explicitly
+    sns.barplot(x="hour", y="count", data=activity_counts, hue="hour", palette="Blues", legend=False)
+
+    plt.xlabel("Hour of the Day")
+    plt.ylabel("Activity Count")
+    plt.title("User Activity Distribution")
+    plt.xticks(range(0, 24))
+
+    # Save the plot file safely
+    plt.savefig(output_path, bbox_inches="tight")
+    plt.close()
+
+    # Verify the file exists after saving
+    if not os.path.exists(output_path):
+        raise FileNotFoundError(f"Plot could not be saved to {output_path}")
+
+    return output_path